• Corpus ID: 237503417

Conditional Synthetic Data Generation for Robust Machine Learning Applications with Limited Pandemic Data

  title={Conditional Synthetic Data Generation for Robust Machine Learning Applications with Limited Pandemic Data},
  author={Hari Prasanna Das and Ryan Tran and Japjot Singh and Xiangyu Yue and Geoffrey H. Tison and Alberto L. Sangiovanni-Vincentelli and Costas J. Spanos},
$\textbf{Background:}$ At the onset of a pandemic, such as COVID-19, data with proper labeling/attributes corresponding to the new disease might be unavailable or sparse. Machine Learning (ML) models trained with the available data, which is limited in quantity and poor in diversity, will often be biased and inaccurate. At the same time, ML algorithms designed to fight pandemics must have good performance and be developed in a time-sensitive manner. To tackle the challenges of limited data, and… 

Figures and Tables from this paper

Expanding New Covid-19 Data with Conditional Generative Adversarial Networks
  • Haneen Majid, K. Ali
  • Computer Science
    Iraqi Journal for Electrical and Electronic Engineering
  • 2022
Conditional Generative Adversarial Networks (CGAN) is developed to produce synthetic images close to real images for the COVID-19 case and traditional augmentation that was used to expand the limited dataset then used to train by Customized deep detection model.
Conditional Synthetic Data Generation for Personal Thermal Comfort Models
It is proposed to implement a state-of-the-art conditional synthetic data generator to generate synthetic data corresponding to the low-frequency classes, and it is shown that the synthetic data generated has a distribution that mimics the real data distribution.
Likelihood Contribution based Multi-scale Architecture for Generative Flows
A novel multi-scale architecture that performs data dependent factorization to decide which dimensions should pass through more flow layers is proposed and a heuristic based on the contribution of each dimension to the total log-likelihood which encodes the importance of the dimensions is introduced.
Graphical Lasso based Cluster Analysis in Energy- Game Theoretic Frameworks
A novel graphical lasso based approach to perform segmentation, by studying the feature correlations in a real-world energy social game dataset and results in characteristic clusters demonstrating different energy usage behaviors are presented.


CDCGen: Cross-Domain Conditional Generation via Normalizing Flows and Adversarial Training
A transfer learningbased framework utilizing normalizing flows, coupled with both maximum-likelihood and adversarial training, that can generate synthetic samples conditioned on them in the target domain by generating non-trivial augmentations via attribute and component transformations.
Synthetic Data Generation for Improved COVID-19 Epidemic Forecasting
This paper proposes a method of strengthening the forecasts from compartmental models by using short term pre-dictions from a curve fitting approach as synthetic data, and discusses the method of fitting this hybrid model in a generalized manner without reliance on region specific data, making this approach easy to adapt.
CovidGAN: Data Augmentation Using Auxiliary Classifier GAN for Improved Covid-19 Detection
This research presents a method to generate synthetic chest X-ray (CXR) images by developing an Auxiliary Classifier Generative Adversarial Network (ACGAN) based model called CovidGAN and demonstrates that the synthetic images produced by this model can be utilized to enhance the performance of CNN for COVID-19 detection.
COVID-19 detection from scarce chest x-ray image data using few-shot deep learning approach
This work has experimented with well-known solutions for data scarcity in deep learning to detect COVID-19 using siamese networks, and proposed a custom few-shot learning approach that was able to achieve 96.4% accuracy an improvement from 83% using baseline models.
Self-training with improved regularization for sample-efficient chest x-ray classification
This work presents a deep learning framework that utilizes a number of key components to enable robust modeling in chest X-ray classification and provides several key insights on the effective use of data augmentation, self-training via distillation and confidence tempering for small data learning in medical imaging.
Few-shot Domain Adaptation by Causal Mechanism Transfer
This work takes the structural equations in causal modeling as an example and proposes a novel DA method, which is shown to be useful both theoretically and experimentally, and can be seen as the first attempt to fully leverage the structural causal models for DA.
COVID-19 CT Image Synthesis With a Conditional Generative Adversarial Network
Experimental results show that the proposed CT image synthesis approach based on a conditional generative adversarial network outperforms other state-of-the-art image synthesis methods with the generated COVID-19 CT images and indicates promising for various machine learning applications including semantic segmentation and classification.
Reinforcement Learning for Optimization of COVID-19 Mitigation policies
This paper presents a novel agent-based pandemic simulator which, unlike traditional models, is able to model fine-grained interactions among people at specific locations in a community and an RL-based methodology for optimizing fine- grained mitigation policies within this simulator.
Meta-DermDiagnosis: Few-Shot Skin Disease Identification using Meta-Learning
This paper trains a neural network on few-shot image classification tasks based on an initial set of class labels / head classes of the distribution, prior to adapting the model for classification on a set of unseen / tail classes, and incorporates Group Equivariant convolutions (G-convolutions) for the Meta-DermDiagnosis network to improve disease identification performance.
COVIDNet-CT: A Tailored Deep Convolutional Neural Network Design for Detection of COVID-19 Cases From Chest CT Images
COVIDNet-CT, a deep convolutional neural network architecture that is tailored for detection of COVID-19 cases from chest CT images via a machine-driven design exploration approach, is introduced and the model and dataset are introduced.